diff --git a/clustering/agg_clustering.py b/clustering/agg_clustering.py
index b8a25e2de281bcc70e242b44dcf3e108f9d37890..6bda725901e758f73e457584c7539b35efad9e35 100644
--- a/clustering/agg_clustering.py
+++ b/clustering/agg_clustering.py
@@ -1,3 +1,4 @@
+from sklearn.cluster import AgglomerativeClustering
 def agglomerative_clustering(reduced_embeddings, n_clusters):
     """
     Applique l'Agglomerative Clustering avec un nombre fixe de clusters.
diff --git a/clustering/mesures_clustering.py b/clustering/mesures_clustering.py
index af1ffdf76328c289e7b40c06d2b9f489c73ef130..9c6c2e749218b218332db86606c90dd49e73145c 100644
--- a/clustering/mesures_clustering.py
+++ b/clustering/mesures_clustering.py
@@ -1,3 +1,4 @@
+from sklearn.metrics import silhouette_score
 def compute_silhouette_scores(reduced_embeddings, labels):
     """
     Calcule les scores de silhouette pour différents nombres de clusters.
diff --git a/projet_imt.py b/projet_imt.py
index 2b34264808cb1c0be5ccbb0a0d2ccb0271f89beb..b31b5d8f56079737c95f8ae9e9587564f2740b3a 100644
--- a/projet_imt.py
+++ b/projet_imt.py
@@ -8,18 +8,14 @@ from utils.reduction_dimesion import *
 
 import pandas as pd
 import numpy as np
-import matplotlib.pyplot as plt
-import ast
 import umap
-from sklearn.cluster import KMeans
 from sklearn.metrics import silhouette_score
 from sklearn.cluster import AgglomerativeClustering
-from sklearn.metrics import silhouette_score
 
 
 def main():
     # lecture du fichier csv
-    df = load .....
+    df = pd.read_csv("/Users/mac/Desktop/topic_modeling/df_user_messages_new.csv")
     # reduction des dimensions
     reduced_embeddings = reduce_embeddings(df, n_components=50, random_state=42)
     # clustering
@@ -29,6 +25,9 @@ def main():
     # afficher les resultats
     messages_par_cluster= afficher_messages_par_cluster(df,labels)
 
+    return messages_par_cluster
+
 
 if __name__ == "__main__":
-    main()
+    messages_par_cluster = main()
+    print(messages_par_cluster)
\ No newline at end of file